The present invention relates to implantable monitoring systems for the continuous in vivo measurement of biochemical substances, and more particularly to improved implantable enzyme-based glucose monitoring systems, small enough to be implanted through the lumen of a catheter or hypodermic needle, that measure the amount and rate of change of glucose in a patient's blood over an extended period of time.
Glucose is an important source of energy in the body and the sole source of energy for the brain. Glucose is stored in the body in the form of glycogen. In a healthy person, the concentration of glucose in the blood is maintained between 0.8 and 1.2 mg/ml by a variety of hormones, principally insulin and glucagon. If the blood-glucose concentration falls below this level neurological and other symptoms may result, such as hypoglycemia. Conversely, if the blood-glucose level is raised above its normal level, the condition of hyperglycemia develops, which is one of the symptoms of diabetes mellitus. The complications associated with both of these disorders, particularly if left uncorrected, can result in patient death. Thus, measuring and maintaining the concentration of glucose in the blood at a proper level is critically important for good health and longevity.
Unfortunately, some individuals are physically unable to maintain the proper level of glucose in their blood. For such individuals, the concentration of glucose in the blood can usually be altered, as required, to maintain health. For example, a shot of insulin can be administered to decrease the patient's blood glucose concentration, or conversely, glucose may be added to the blood, either directly, as through injection or administration of an intravenous (IV) solution, or indirectly, as through ingestion of certain foods or drinks.
Before a patient's glucose concentration can be properly adjusted, however, a determination must be made as to what the current blood glucose concentration is and whether that concentration is increasing or decreasing. Many implantable glucose monitoring systems have been described that are designed to provide continuous measurement of a patient's blood glucose concentration. See for example, U.S. Pat. Nos. 3,539,455; 3,542,662; 4,484,987; 4,650,547; 4,671,288; 4,703,756; 4,890,620; 5,165,407; and 5,190,041. Most of these systems are based on the "enzyme electrode" principle where an enzymatic reaction, involving glucose oxidase, is combined with an electrochemical sensor, to measure either oxygen or hydrogen peroxide, and used to determine the concentration of glucose in a patient's blood.
Generally, enzyme-based glucose monitoring systems, whether implantable or not, use glucose oxidase to convert glucose and oxygen to gluconic acid and hydrogen peroxide (H.sub.2 O.sub.2). An electrochemical oxygen detector is then employed to measure the concentration of remaining oxygen after reaction of the glucose; thereby providing an inverse measurement of the blood glucose concentration. A second enzyme, catalase, is optionally included with the glucose oxidase to catalyze the decomposition of the hydrogen peroxide to water, in order to prevent interference in the measurements from the H.sub.2 O.sub.2.
Thus, this system of measuring glucose requires that glucose be the limiting reagent of the enzymatic reaction. Where the system is to be used in vivo, this requirement can, and often does, pose a serious problem because, on a molar basis, the concentration of free oxygen in vivo is typically much less than that of glucose. This "oxygen deficit" prevents the exhaustion of glucose in the area of the enzymatic portion of the system and thus, results in an inaccurate determination of glucose concentration. Further, such an oxygen deficit can contribute to other performance related problems for the sensor assembly, including diminished sensor responsiveness and undesirable electrode sensitivity. See for example, the discussion in Gough et al., in Two-Dimensional Enzyme Electrode Sensor for Glucose, Vol 57 Analytical Chemistry pp 2351 et seq. (1985), incorporated by reference herein.
Attempts to solve the oxygen deficit problem, associated with in vivo glucose monitoring systems have previously been presented and are primarily based upon either reduction of the enzyme catalytic activity or regulation of the diffusion of glucose and oxygen through the use of specialized membranes. See for example, U.S. Pat. No. 4,484,987, Gough, D., hereby incorporated by reference, in its entirety. These solutions, however, have their own disadvantages. For example, reduction of the enzymatic activity of the monitoring system requires either a reduced concentration of enzyme or a thinner layer of active enzyme, both of which tend to shorten the useful life of the enzymatic sensor by reducing the amount of useful enzyme. Alternatively, a longer thinner layer of enzyme can be employed within the sensor assembly in order to locally reduce the enzymatic activity without loss of useful life to the sensor, but this tends to slow the responsiveness of the sensor and/or requires a larger (i.e. longer) sensor, which are both undesirable.
Use of specialized membranes to control the diffusion of glucose and oxygen into the sensor assembly, can also present problems. For example, as discussed in U.S. Pat. No. 5,322,063 issued to Allen et al. and incorporated by reference herein, in its entirety, controlling the diffusion of the glucose and oxygen through the use of specialized membranes can lead to slower responsiveness of the sensor and/or unintentional poisoning of the sensor and electrodes caused by migration of undesirable substances through the specialized membranes to the sensor. In particular, to the degree the specialized membrane comprises a hybrid of two different membrane types, having numerous junctions between two or more disparate membranes, concerns arise as to the integrity of such junctions, the appropriate ratio of the two membranes to one another and the appropriate configuration of the hybrid (i.e., for example, "islands" of one membrane type within and each surrounded by the other membrane type or alternating "stripes" of membrane types).
An additional concern that arises when a monitoring system is intended for use within the body of a patient, especially where it is to be used long term, relates to the biocompatibility of the system. The protective mechanisms of the body attempt to shield the body from the invasion of the monitoring system which is perceived as an unwanted foreign object. These protective mechanisms include, for example, encapsulation of the foreign object by the growth of isolating tissue and coagulation of blood on and around the foreign object. Obviously, encapsulation of and/or blood coagulation around all or part of the implantable sensor can significantly reduce or completely terminate the functionality of the device.
Often, the exterior surface of implantable monitoring systems are composed of silicone rubber, which is a reasonably biocompatable material. However, silicone rubber has been shown to induce thrombosis and encapsulation when compared to living endothelium. Many different approaches have been utilized to enhance the biocompatibility of silicone rubber and similar polymeric materials. However, presently, it is unclear exactly what the relationships of surface chemistry and morphology are to blood/body compatibility. This could be due to several factors. First, polymer surfaces often are not well characterized. Second, additives, processing aids and the like may have migrated to or may have been left behind at the polymer surface; thereby contributing to the unpredictable nature of the surface. Furthermore, even if the polymer is pure, it can have varying surface configurations. The polymer surface may not be homogeneous, or coatings, if any, may not be uniformly applied, or the surface, itself may have developed cracks, all of which can contribute to the incompatibility of the surface with the body.
With some degree of success, polyethylene glycol (PEG) has been used to coat polymeric surfaces in order to repel proteins from the surface after implantation. Similarly, heparin has been either covalently bound or ionically bound to polymeric surfaces in order to prevent blood coagulation thereabout after implantation. However, ionically bound heparin is gradually released from the implant surface providing only short-term anticoagulant effect, and covalently bound heparin, while remaining bound to the surface for longer periods of time, is not as effective an anticoagulant. Thus, where the object to be implanted is intended to remain for a long period of time, the anticoagulant of choice is covalently bound heparin.
While finding solutions to the problem of biocompatibility can be quite difficult, since proper operation of the sensor requires that certain membranes be in contact with the patient's blood, still what is needed are membranes that are biocompatible yet maintain their functionality or that may be treated, for example by application of an appropriate coating, to possess these characteristics. Additionally, an enzyme-based glucose monitoring system designed to protect against the problem of oxygen deficit while providing quick, accurate and continuous glucose concentration readings over a long term is desirable.